Academic journal article
By Howard, Robert M.; Nixon, David C.
Political Research Quarterly , Vol. 55, No. 4
While there have been numerous studies demonstrating bureaucratic control of administrative and regulatory agencies, many argue that the Internal Revenue Service is an agency out of control, and one not subject to political constraints. However, some recent studies have shown that the IRS is subject to some political control in shifting policy between the often competing concerns of efficiency and fairness. We extend these studies to examine judicial control of the IRS. Examining cross sectional time series data from 1960 until 1988, we use regression with panel corrected standard errors to show that, while efficiency is always very important, the IRS shifts audits between the wealthy and the less affluent in response to the prevailing median ideology of the relevant federal court of appeals. As the median appeals court judge becomes more liberal, the IRS shifts its audits in that region in favor of equity by reducing the audits on the poor and increasing the audits of the more affluent. As the median appeals court judge becomes more conservative, the IRS shifts the audits in that region in favor of greater efficiency by increasing the audits of those in the lower economic strata. Courts provide an additional measure of control of bureaucratic behavior.
Democratic control of bureaucracies has been the subject of numerous empirical and normative examinations. Many theorists have focused on normative goals of how to control bureaucracy (e.g., Gormley 1989) while other scholars have empirically demonstrated institutional constraint of bureaucratic behavior, (e.g., Fiorina 1979) showing that this "headless fourth branch of government" is in fact subject to democratic control.
However, in contrast to these bureaucratic control studies, many citizens, journalists, elected representatives, and scholars argue that the Internal Revenue Service, arguably one of the largest and most powerful of all governmental agencies, is not subject to the same democratic pressures and constraints as other executive agencies. These scholars, journalists and officials often claim that the IRS does not carry out the policy wishes of the president and ignores congressional and court control (see, e.g., Long 1980; Burnham 1989; MacDonald 1994).
An examination of IRS audit data provides some reinforcement to this position. For a national agency addressing a national problem of collection and enforcement, there is significant variation from state to state that seems driven by the idiosyncrasies of IRS personnel. For most of the post World War II period, The IRS operated out of 60 districts within state boundaries in a larger set of 7 national regions, and currently the IRS audits out of 33 districts in 4 national regions. The states, districts, and regions all exhibit significant variance in audit rates. For example, one journalist notes that a citizen of Nevada, which has a statewide audit rate of 1.8 percent, is seven times as likely to be audited than is a citizen of Wisconsin with a statewide audit rate of 0.24 percent (MacDonald 1994: 132). A citizen of the Middle Atlantic region has a 20 percent greater probability of getting audited than a citizen of the Midwest. While some variation is to be expected because of differing local economic conditions, popular accounts seem to suggest that the variation is somehow evidence of a lack of political control of the agency
Contrary to these assertions, we argue that the IRS operates the same as other agencies. Although the agency seeks efficiency in revenue collection, the balance between efficiency and equity is subject to the same controls by the executive, legislative, and judicial branches as other agencies. One explanation for the regional variation is the differing ideologies and policy preferences of the various federal appellate courts. In this article, we investigate the state level variation in audits of wealthy versus poor taxpayers over the 28-year period from 1960 to 1988. …